Satellites

The Satellites section answers “which bird can I work, and when?” for your grid. It predicts amateur-satellite passes over your location, schedules your ★ birds 48 h out, plots each pass, lists the working frequencies, and — if you have a rotator — can auto-track a bird across the sky through a pass.

Satellites is an opt-in section. Turn it on in the first-run wizard or in Settings ▸ Appearance ▸ Features. It needs your grid set in Settings ▸ Station to compute passes.

The tour

The section is laid out as a pass console: everything you need to work a pass is on one screen, and you scroll only for the things you use between passes. From a 1200×750 window upward that is literal — the dome, its rise/set readout, the globe, the pass timeline and the log strip down to its Log button are all visible at once. At 1024×768, the smallest window Nexus supports, the last inch of the log strip sits just under the fold; see what fits at which window size below.

Across the top, the arm bar says what is armed and what it is driving — the bird, which rig is bound, the readiness gates (pass, rotor, transponder, Doppler, elements), the ■ that stops the track and the ✕ that closes the bird.

On the left, the planning column. Next and Best 24 h lead with the two soonest workable passes and the two best ones, each with a ▶ Work this pass. Under them the 48 h schedule for your ★ birds scrolls inside its own box — the rest of the page does not move when you scroll it — with the “other birds overhead” disclosure pinned above it so it never scrolls away. At the bottom sit the frequencies (the live Doppler-corrected dial and where you are inside the passband) and the transponder chooser. Those two stay put: picking a transponder is the most consequential thing you do here, so it is never behind a scroll.

On the right, the pass column. The sky dome draws the pass in az/el with the AOS and LOS bearings on their marks, and the ground-track globe sits beside it showing where the bird is over the earth and whose grids its footprint crosses. Under them the pass timeline, then the log strip, then your Birds catalog — which is the one thing deliberately below the fold.

What fits at which window size

The console does not shrink its contents to fit; it tells you where the fold is. Measured with a full 42-pass schedule, mid-pass, with a track armed:

Window	Schedule rows	The whole pass set on screen?
1024×768 (the supported minimum)	5	Almost — the Log button is 6 px under the fold, the note under it 36 px
1280×800	8	Yes
1366×768	7	Yes
1600×900	7	Yes
1920×1080	16	Yes
1200×1390 (tall)	24	Yes
3440×1440 (ultrawide)	30	Yes

The schedule is the only thing that grows, so every pixel the rest of the console gives back becomes another row — and the “first ten lines” worth of schedule arrives at about 900 px of window height.

Two more honest notes about the 1024×768 minimum. The pass column scrolls by about an inch to reach the bottom of the log strip — it already scrolls to reach the Birds catalog, so this is the same flick of the wheel. And the schedule table is wider than its column there, so it scrolls sideways inside its own box — about an inch of it at 1024 wide, a few pixels at 1200–1280, and none at 1366 and up.

Opening the transponder chooser’s show all N costs schedule rows — the frequencies panel and the schedule share the column, and that is the trade the layout is built on. It stops growing at a little over half the column and scrolls inside itself after that, so it can never take the schedule away.

Frequencies for each bird are listed so you know where to listen and where to transmit.

You can also drop a Satellite Passes pane into the Connect grid for an at-a-glance next-passes list beside the map, and turn on the Satellites (amateur) map layer to watch the birds move in real time.

Core workflows

Star your favorites

Click the ⭐ on a bird to favorite it. The 48 h schedule is your ★ birds — everything else overhead sits behind the Other birds overhead disclosure above it. The ISS is the easiest first target — star it and every ISS pass over you turns up in the schedule.

Set a pass alarm

Arm an alarm on a pass and Nexus reminds you before AOS so you don’t miss it. (For the loud, repeating “they’re on the air” style of alarm, see the DXpedition wake-me alarm — the same alarm machinery.)

Tune around the passband — and get back on the bird

Working a linear bird means chasing a station across the transponder, so turn the VFO and Nexus follows: it takes your new dial as your position in the passband and moves your uplink to match (mirrored, if the transponder inverts). Doppler keeps correcting around wherever you put yourself.

Tune outside the passband and that stops — you’ve left the transponder as far as Nexus can tell, which is the right call, because the alternative is dragging somebody’s uplink to a passband edge because they QSY’d to 20 m. Lock on, on the Dial line under the bird’s name, puts you back: it re-runs the transponder pick you already made, so the radio, the band and the mode all come with it, and you land in the middle of the passband again.

It is there from the moment you pick a transponder — that pick tunes the radio straight away, so the dial is live long before you arm a pass, and the way back is live with it. It sits with the line that names the rig, and it is there whether the pass is armed or not and whether or not the bird is up. The one state it is absent in is the one where it would have to guess: with no transponder picked there is nothing to put you back onto, and choosing one for you would be choosing your uplink.

Log the contact without leaving the pass

The log strip sits in the pass column under the sky dome and the pass timeline — the same log strip the Phone and CW cockpits use, with the same callbook lookup, the same recall card and the same prior-contact history. It is there whether or not a pass is armed, and it stays there after the bird sets, so you can catch up on a contact once your hands are free.

Nothing you do in the section can take a half-typed contact away from you. Closing the bird with ✕ or Escape, clicking a different bird, arming a pass, AOS arriving, or Nexus losing its connection to SatNOGS mid-pass all leave the form exactly as you left it. (Through 0.28.1 they did not: the strip lived inside the bird’s detail card, so any of those wiped what you had typed.)

Type the call and press Enter twice. On a call the strip hasn’t seen yet the first Enter runs the callbook lookup and fills the name and QTH; the second one logs. (Once the name is filled, one Enter logs.) The band, frequency, mode and time come from where you already are, and the report defaults to the one for that mode. Working a bird from a rig Nexus isn’t connected to? Open Log a contact from another radio in the strip and set the band, frequency, mode and UTC time by hand.

The grid goes in the Grid box, beside the two reports. Satellite work is grid-for-grid, so type the square he passed you — it sits with the reports because that is when you hear it. It is the one field the callbook regularly gets wrong for satellite work — a rover or a portable operator gives you where he is, and his callbook says where he lives — so the box wins: a lookup fills it only while it is blank and never overwrites what you typed. Clearing it and running the lookup again gets you the callbook’s square back.

Nexus takes a 4-, 6- or 8-character locator: EN52, EN52XA or EN52XA25 — every length ADIF’s grid field carries, so all three upload. Case doesn’t matter, it uppercases as you type. Anything else — a half-typed EN5, two squares in one box — is refused: the Log button goes dead and the line above it says which forms it takes, until you fix the square or clear the box. It refuses rather than quietly dropping what you typed, because a QSO record is permanent and a wrong square is not a missing grid but a grid credited to a square you never worked. A blank Grid box is not an error.

A callbook answer that is not a locator never lands in the box at all — nothing you did not type can put the Log button out of reach in the middle of a pass.

The box is in this section only for now. The Phone and CW strips do not have one yet: it costs each of them a wrapped line, and there was no room to spend on that here. It is still on the table.

There is no park row here. POTA and SOTA are a terrestrial exchange, and the Phone and CW strips still carry the picker and the park search. This section asks the same strip for a satellite exchange, so that row isn’t built here — one fewer thing between the Doppler readout and the sky dome in this column.

It logs an ordinary contact, not a satellite contact. This is worth being plain about, because it decides whether a contact can ever earn satellite credit. LoTW recognises a satellite QSO by two ADIF fields:

• PROP_MODE=SAT — the propagation mode.
• SAT_NAME — the satellite, spelled the way LoTW spells it (AO-7, not AO7).

Nexus writes neither, for any contact it logs. Getting SAT_NAME right means resolving the bird’s designator, and ARRL is explicit that a name LoTW does not recognise gets the data rejected — so a guess is worse than nothing, and a logged QSO is permanent. That resolution is not built yet. Until it is, if you want satellite credit you have to add the two fields yourself: set them in whatever logger you upload from, or add them to the ADIF before you sign it. Any record that already carries them — a foreign import, or one you repaired — keeps them: Nexus writes them out on export and reads them back on import untouched.

It has to be both fields, and that is why Nexus writes neither rather than the easy one. PROP_MODE=SAT on its own would be trivial to write and it would be true — but TQSL will not sign a contact whose propagation mode is SAT when it names no satellite (“PROP_MODE = ‘SAT’ but no SAT_NAME”), just as it will not sign one naming a satellite it doesn’t know. Either way the contact never reaches LoTW, so it earns nothing at all: not the satellite credit you were after, and not the DXCC or WAS credit an ordinary untagged upload does earn. A half-tag costs you the whole QSO. When you add the fields by hand, add both.

Three things this strip does not do yet

None of the three is a decision that satellite work should stay this way. They are the price of dropping the Phone/CW log strip in unchanged rather than building a satellite-aware one, and each is meant to be closed.

Your satellite grids land in the wrong place — in Nexus and at ARRL. Nexus decides “was this a satellite QSO?” from PROP_MODE=SAT in the record. (LoTW asks for more than Nexus does: it wants that field and the satellite’s name.) With nothing writing it, a contact logged here counts toward neither the Satellite VUCC totals on the Awards screen nor the satellite needs board that ranks which pass is worth chasing.

Where the grid goes instead depends on the band you were listening on, and neither answer is right:

• Outside 160 m – 2 m it lands nowhere. Nexus keeps per-band grid counts for 160 m through 2 m only — so 2200 m and 630 m below that range, and 1.25 m, 70 cm, 23 cm, 13 cm, 9 cm, 5 cm, 3 cm and 1.2 cm above it, have no slot of their own, and the satellite bucket was the only home they had.
• On 160 m through 2 m it lands in the wrong bucket. For satellite work that means 2 m — the downlink of every U/V bird, the Fox-1 satellites (AO-85, AO-91, AO-92) and AO-7 on mode B — and 10 m, where AO-7’s mode A comes down. So this is the ordinary case, not a corner. The grid is counted toward your terrestrial VUCC for that band, which is a grid ARRL’s rules say a satellite contact does not earn, and LoTW files the untagged upload the same way. If you chase VUCC, this one matters: add both fields before you sign.

The split is by band, not by whether the band is named in metres: 1.25 m is a metre band and it falls out with the centimetre ones.

Adding the two fields by hand puts the contact where it belongs in both counts: edit the ADIF with Nexus closed, exactly as below, and the totals pick it up the next time Nexus starts.

During Field Day, this strip logs to the ordinary log, not the contest log. The Phone and CW strips switch to the Field Day log while a session is running; this one is not wired to Field Day yet, so a satellite contact made during FD goes into your general log and scores the club nothing. Until it is wired, log satellite contacts made during Field Day from the Phone or CW cockpit while you are still on the bird. Catching up afterwards does not work cleanly: the FD log stamps every contact with the band the radio is on at the moment you type it, so a 70 cm pass entered later goes into the contest log — and out to N1MM or N3FJP — on whatever band you have since moved to.

The recorded MODE comes from your sideband, so it is wrong on a data mode. The strip writes FM, CW, AM or SSB, chosen from the mode the rig is commanded to. That is right for voice and CW work. But the Satellites section is also reachable on the digital tiers, and there the sideband is not the mode: on FT8, FT4, Q65, JT65, MSK144, WSPR and FST4 every channel commands USB, so the contact records MODE=SSB; on Tempo’s three FM simplex channels (2 m, 1.25 m, 70 cm) it records MODE=FM. Either way the record names a voice mode for a contact you made on a data mode.

Until the strip is tier-aware there are two fixes, and only one of them covers your tier:

• On FT8 or FT4, open Log a contact from another radio in the strip and pick the mode there before you log.
• On Q65, JT65, MSK144, WSPR, FST4 or Tempo, that picker has no entry for your mode — it offers SSB, FM, AM, CW, RTTY, FT8 and FT4 and nothing else. Log the contact, then open the Logbook, edit the record and type the mode into its Mode field, which takes any text.

If you ran 0.24.0 through 0.27.x, check your log. In those versions a contact logged while a transponder was held picked up PROP_MODE=SAT and a SAT_NAME taken from the catalog name of the bird — “SAUDISAT 1C (SO-50)”, not “SO-50”. LoTW does not recognise those, and the hold is only handed back when the pass ends (a transponder picked without arming a pass is never handed back at all), so ordinary contacts made afterwards were tagged too. Nexus no longer writes any of it.

Existing records are left exactly as they are. Nexus will not rewrite contacts you already logged: some of them really were satellite QSOs that want the name corrected, and some were terrestrial contacts that want the tag gone, and nothing in the record tells the two apart — only you know which pass you were actually on. To find them, your general log is a plain ADIF file (~/.config/tempo/log.adi, or %APPDATA%\tempo\log.adi on Windows): search it for SAT_NAME. Fix them there, with Nexus closed — correct the name, or delete both fields from the record. There is no way to do it from inside Nexus: the logbook’s edit form does not carry these two fields, and an edit that leaves them blank deliberately preserves what is stored, so that an ordinary busted-call fix cannot silently strip a satellite tag off a record that earned it.

Pin the radio a pass uses

On a multi-radio station the readiness rail names the rig a pick routed to, and the band and mode class it routed on. If two rigs cover the downlink, the pick goes wherever your routing says — which is usually what you want, and sometimes not. 🔓 pin this radio holds the pass on the radio you are on: it stays 🔒 pinned until you click it again, and no pick hands the bird to another rig meanwhile. It is the same switch as the 🔒 beside the radio selector in the top bar, put where you are working the pass. Pinning does not re-tune anything — it decides where the next pick lands.

Auto-track with a rotator

1. Configure your rotator in Settings ▸ Radio ▸ Rig Control — pick your model and its COM port and Nexus runs the control daemon for you. No hardware? Pick the Dummy (testing) model, or run rotctld -m 1 and point Nexus at 127.0.0.1:4533 to watch it work.
2. Arm rotor track on a pass. Nexus slews the rotator to follow the bird across the sky through the pass; the compass shows the track, with °T and °M side by side, and a STOP control halts it.

The section is read-only until you arm a track — it won’t touch your rotator on its own.

If the rotator stops answering mid-pass, Nexus gives up on it rather than hammering it for the rest of the pass — and gives up on only it. The pass carries on: your transponder stays picked, Doppler keeps correcting the radio, and the sky dome keeps showing where the bird is so you can turn the antenna yourself. The track says so plainly (the badge and the readiness rail both read “rotor stopped answering”) and stops showing commanded angles, because it has stopped commanding anything. The rotor strip in a cockpit header keeps naming the bird and the ■ that stops the track, beside the dim “ROTOR —” for the mast.

At LOS that pass still sends one stop, in case the controller came back — but it will not park or go to ready, even if you configured one. The pointing was handed to you, so the antenna stays where you left it.

Honest limits

• Passes are computed for your grid — set your Maidenhead locator first or the predictions can’t run.
• The bird list is not everything in orbit — it is the amateur population: satellites with an amateur transmitter on record. Around 430 birds are listed and around 367 of those carry current orbital elements.
• Rotor auto-track drives an az/el rotator through Hamlib rotctld (elevation is followed through the pass; an azimuth-only rotator is detected automatically and driven in azimuth alone); test it with the Dummy model before you trust it on real hardware.

Where does the bird list come from, and what does a bird’s status mean?

The list is derived, not copied. It starts from the SatNOGS database — the community record of which satellites carry which transmitters — and keeps every satellite with an amateur transmitter: one SatNOGS labels Amateur, or one transmitting in an ITU amateur-satellite allocation (the band test is what keeps SO-50, whose transmitters are all filed as “Unknown”). Orbital elements for those birds are then assembled from three sources, freshest epoch winning: CelesTrak’s amateur group, CelesTrak’s satnogs group, and the SatNOGS element service. That last one matters — it is the only source for a bird still catalogued under a placeholder number, and for birds CelesTrak has no elements for at all.

The list is rebuilt every six hours by the project’s mirror, so a bird going on or offline reaches you within six hours of SatNOGS recording it — no app update needed. Each bird’s status rides with it:

• alive — in orbit, with something amateur transmitting. These are the birds that carry elements and appear on the map, in the schedule and in the pass list.
• alive but silent — in orbit, but the catalog lists no working amateur transmitter any more. The pass geometry would still be real; there is nothing to work on it.
• dead — reported silent.
• re-entered — gone. Kept in the list for six months after re-entry, so a favorite that stops working has a row that says why, then dropped.
• pre-launch — on record but not yet deployed. Nothing to work yet.

Only alive birds carry elements, so a bird in any other state shows in the list with its status and “no elements” rather than a position. Your ★ stays put either way — a bird that dies never vanishes out from under its star, and search reaches the whole catalog so you can always find it to unstar it.

How current are the elements?

Every prediction runs on orbital elements (TLEs), and elements decay: a fresh set predicts a pass to the second, an old one drifts — and pointing and Doppler drift with it. Nexus keeps the elements current for you and tells you plainly when it can’t:

• Where they come from. Elements refresh in the background a few times a day from the project’s mirror described above (CelesTrak data courtesy of Dr. T.S. Kelso; population and status data from SatNOGS / Libre Space Foundation, CC BY-SA 4.0); the mirror exists so a fleet of installs never hammers the sources. If the mirror itself is unreachable for a day, Nexus falls back to fetching CelesTrak’s amateur group directly — a shorter list of 97, with every other bird keeping its row marked “no elements” until the mirror is back. The Satellites section and Settings ▸ Orbital elements show the age, fetch time and source. Update now forces a refresh; Import from file loads a downloaded TLE/keps file — the path for offline shacks and brand-new launches no source carries yet.
• Past 14 days the age is badged stale, and arming a pass asks first — refresh right there, or arm anyway with your eyes open.
• Past 30 days SGP4 accuracy is genuinely gone, so anything that would move the radio or rotator refuses — naming the bird and the age — rather than drive the antenna off a fiction.
• A pass runs on the elements it armed with (a pass is minutes long; a mid-pass swap would jump the antenna). The readiness rail shows the age of the frozen set.
• Renames don’t orphan your stars. CelesTrak occasionally renames a bird; your ★s, alarms and schedule keep working because Nexus remembers the catalog number behind each name.
• A bad or empty download never replaces a good cache, and nothing waits on the network — Nexus serves the best elements it has and refreshes behind you.

Related guides

• Connect — map + propagation (Satellite Passes pane, live map layer)
• SSTV (ISS SSTV auto-arm: at AOS Nexus tunes 145.800 FM and arms the decoder for you)
• Settings reference (rotator setup)
• DXpeditions (the same alarm machinery)